Forests and Climate Change Working Paper 4
Choosing a forest definition
for the Clean Development Mechanism
)$26
FORESTS AND CLIMATE CHANGE WORKING PAPER 4
CHOOSING A FOREST DEFINITION
FOR THE CLEAN DEVELOPMENT MECHANISM
Till Neeff
Heiner von Luepke
Dieter Schoene
Table of Contents
Abstract .................................................................................................................................................. 3
Forest, afforestation and reforestation as defined for the Clean Development Mechanism .......... 3
Remaining ambiguities of definitions .................................................................................... 4
Considering existing national definitions............................................................................... 5
Choosing parameter values .................................................................................................................. 6
Furthering national policy objectives with the CDM............................................................. 7
Accommodating prior site conditions .................................................................................... 7
Facilitating project types ........................................................................................................ 9
Integrating prior site conditions and project types ............................................................... 10
What parameter values have NAI countries selected up to the present? ............................. 12
Conclusions .......................................................................................................................................... 13
References ............................................................................................................................................ 15
Appendix I............................................................................................................................................ 17
Appendix II .......................................................................................................................................... 18
Agroforestry practices in small-scale A&R ...................................................................................... 18
1
Abstract
Developing countries must define “forest” before they can host afforestation and reforestation projects under the
Clean Development Mechanism of the Kyoto Protocol. To do so, they must chose country-specific values from a
range provided in the Marrakech Accords for minimum area, crown cover and tree height. Good practice
involves choosing also a minimum strip width. Definitions in the Marrakech Accords and in the 2003 IPCC
Good Practice Guidance leave some ambiguities.
Existing country definitions of “forest” do not contain all or, sometimes, any quantitative parameters; simply
adopting them for the CDM is not an option. Therefore, all developing countries vying for forestry projects
under the CDM will have to choose parameter values. Only few have done so up to now.
Parameter values affect land eligibility for the CDM, feasibility of project types and match with national policy
goals. Choosing parameter values that maximize eligibility and feasibility may conflict with those that match
economic and socio-environmental expectations. Given a multitude of goals and project types, a large spectrum
of sites, current land-uses and ownership patterns, there are no generally valid, optimal choices. A stepwise
decision process and simple graphical techniques provided here may aid NAI countries in choosing optimal
parameter values for the forest definition.
Forest, afforestation and reforestation as defined for the Clean
Development Mechanism
The Kyoto Protocol limits carbon sequestration projects under the Clean Development
Mechanism (CDM) during the first Commitment Period to afforestation and reforestation
(A&R). The Marrakech Accords define:
“Afforestation” is the direct human-induced conversion of land that has not been forested for a period of at least
50 years to forested land through planting, seeding and/or the human-induced promotion of natural seed sources;
“Reforestation” is the direct human-induced conversion of non-forested land to forested land through planting,
seeding and/or the human-induced promotion of natural seed sources, on land that was forested but that has been
converted to non-forested land. For the first commitment period, reforestation activities will be limited to
reforestation occurring on those lands that did not contain forest on 31 December 1989
[FCCC/CP/2001/13/Add.1].
Most projects are expected to involve reforestation (Dutschke 2002), necessitating prove that
forest land was converted to non-forest land use before 1990. Therefore, the definition of
“forest” is crucial for the prior and contemplated future state of the land within the
prospective project boundary.
For the CDM, developing countries must choose the parameter values from the ranges1:
“Forest” is a minimum area of land of 0.05-1.0 hectares with tree crown cover (or equivalent stocking level) of
more than 10-30 per cent with trees with the potential to reach a minimum height of 2-5 metres at maturity in
situ. A forest may consist either of closed forest formations where trees of various storeys and undergrowth
cover a high proportion of the ground or open forest. Young natural stands and all plantations which have yet to
reach a crown density of 10-30 per cent or tree height of 2-5 meters are included under forest, as are areas
normally forming part of the forest area which are temporarily unstocked as a result of human intervention such
as harvesting or natural causes but which are expected to revert to forest; [FCCC/CP/2001/13/Add.1]
Alternate values for definition parameters can have significant impacts. When FAO redefined
forest between the 1990 and 2000 Forest Resource Assessments, reducing minimum height
1
The IPCC Good Practice Guidance stipulates selecting in addition a minimum strip width for “linear forests”.
(IPCC, 2003). FAO defines a threshold of 20 m. However, since the Marrakech Accords only list crown cover,
tree height and area, the UNFCCC records only these parameters.
3
from 7 to 5m, minimum area from 1.0 to 0.5 ha and crown cover from 20% in developed and
10% in developing countries to a uniform 10%, global forest area increased by 300 Million ha
or approximately 10% from that cause (FAO, 2000a). The selection of the forest definition is
binding only for the first Commitment Period, 2008-2012. Even countries with very diverse
ecosystems must choose one set of parameters for their entire territory2.
“Forest” is not defined by land use, but rather by the current or expected physical properties
of vegetation cover (Verchot et al. 2005). Even land cover systems that are not intuitively
perceived as forests may qualify, e.g. orchards, oil palms, trees planted as shelter belts in
fields, along riversides or highways, as well as urban tree plantings (Dutschke 2002).
Therefore, a wealth of possible projects arises beyond usual forest plantations.
As of September 15, 2006, nineteen NAI countries had reported their forest definitions. In
many other countries, discussions have begun. Given that CDM projects can start anytime and
that competition among host countries will be intense, countries must decide as soon as
possible.
Remaining ambiguities of definitions
The definition of a tree: The Marrakech Accords do not define a tree. According to FAO, a
tree is: “a woody perennial with a single main stem, or, in the case of coppice, with several
stems, having a more or less definite crown; includes bamboos, palms and other woody plants
meeting the above criteria” (FAO 2005). Banana plantations would not qualify as forests,
even though the constraints of minimum height, area, crown cover might be satisfied.
Conversely, oil-palm plantations, bamboos, or fruit orchards may be eligible if they match the
definition.
Potential vegetative cover: The forest definition refers to potential features of trees, leaving
space for a wide range of possible interpretations. Even temporarily unstocked or understocked areas could be forests, as long as they are “expected” to attain the thresholds for
crown cover or tree height at maturity. The definition probably does not refer to long-term
potential natural vegetation, since this would render most lands forests. On the other hand,
continuous suppression of natural regeneration due to grazing, or the reduction in size of trees
due to pruning in some agroforestry systems, would render these permanently “non-forest”.
Trees established on a degrading site may also never reach the predicted threshold for height.
Reforestation occurs only on “lands that did not contain forest on 31 December 1989”. If
forests regenerated naturally after 1990 on such land, a potential must already have existed
before 1990.
Measures of stand height and crown cover: The MA do not define which measure of tree or
stand height among those commonly used in forestry, e.g. mean height, top height, is to be
applied. The differences between these measures can reach several meters (Prodan, 1965).
Crown cover, as used in the MA forest definition, is correctly defined as the percentage of the
ground covered by a vertical projection of the outermost perimeter of the natural spread of the
foliage of plants, which cannot exceed 100%. It is synonymous with canopy cover (IPCC,
2
In some countries where landscape is highly variable, this might not be adequate (Dutschke 2002). Therefore,
countries have discussed the introduction of biome-specific forest definitions, however, there are technical
challenges (Rakonczay 2002). For now, the issue was deferred to be considered as an option for later CPs only
(FCCC/SBSTA/2004/L.26).
4
2003). However, crown density as applied in the MA forest definition, or crown closure as
used at times in the IPCC GPG refer to different concepts (Helms, 1998).
Size of spatial assessment units: The MA contain minimum forest area as a parameter, but
for developing countries no spatial assessment unit for crown cover. Consistency with the
chosen minimum forest area is not required3. By choosing the assessment area sufficiently
large, an area could conceivably contain a significant number of trees without being
considered a forest, as long as the mean crown cover remains below the minimum value. Preexisting patches of trees could be cleared before establishing a project without formally
undertaking deforestation (Neeff 2005).
Human-induced promotion of natural seed sources: While planting and seeding represent
accepted forestry terminology, the term “human–induced promotion of natural seed sources”
does not. It could be misunderstood and appears to reflect intended ambiguity about the
eligibility under the Kyoto Protocol of natural succession to trees. Very many tree species
propagate, naturally or assisted by humans, not from seed, but by rhizomes, runners, suckers
or rooting of living branches in contact with the soil (Bellefontaine, 2005). Taking the
definition literally, any assisted natural succession to trees on non-forest lands by these
mechanisms would not constitute afforestation or reforestation. FAO defines “natural
expansion of forest” as “expansion of forests through natural succession on land that, until
then, was under another land use”. In using this term and requiring human intervention, e.g.
by fencing against grazing or browsing or via site preparation, the intent of the KP would be
expressed unambiguously.
Temporarily unstocked lands: Temporarily unstocked areas form part of a forest. However,
“temporary” remains undefined in the Marrakech Accords. FAO has at times considered an
unstocked area non-forest, if it is not expected to revert to trees within a period of 10 years
(FAO, 2000).
Considering existing national definitions
In many cases, national forest definitions reflect specific biophysical and social conditions of
countries; they are anchored in history, law and forestry practice. Applying such a national
definition to the CDM would be simple, consistent, assure the legal status of project areas and
link them to existing databases. However, while some national forest definitions resemble or
equal those of the Kyoto Protocol or FAO4, most countries’ definitions do not quantify some
or any parameters. Therefore, national definitions can not be transposed in most instances.
Types of national forest definitions in NAI countries
Of the 122 NAI countries which also reported their definitions to FAO, 44 countries employ
functional definitions only, referring to ecological zones, forest types and land use. Forty
countries used at least one quantitative threshold to define forest nationally. Twenty-one
countries applied the FAO forest definition. Seventeen did not define forests nationally and
none of the NAI countries had a complete set of forest parameter values that could directly be
3
The Annex I countries have to choose and report spatial assessment units for application in the context of
3.3/3.4 activities (22/CP.7), which have to be consistent with the minimum forest area (4.2.2.3, GPG-LULUCF).
4
FAO definitions and parameter values in this text are used merely as a reference, not as a recommendation.
5
used for the CDM5. Therefore, all NAI countries opting for forestry CDM projects will have
to define a set of values.
For the Global Forest Resource Assessment 2005, most widely used parameters were, in
descending order, minimum canopy cover, minimum height and minimum area. Values used
most often correspond to values employed in the FAO forest definition i.e., minimum height 5
m, canopy cover 10%, and minimum area 0.5 ha (Figure 1).
12
16
FAO
10
8
6
4
Number of countries
10
12
Number of countries
Number of countries
14
8
6
FAO
4
6
FAO
4
2
2
2
0
0
20
40
60
80
0
0
crown cover (%)
2
4
6
8
minimum height(m)
10
12
0
0
1
2
3
4
5
6
7
8
9
10 11
minimum area (ha)
Figure 1: Distribution of parameters values for crown cover, height and area in national definitions of NAIcountries and FAO parameters as reference.
Choosing parameter values
The following ten steps may facilitate the choice of parameter values for the forest definition:
1. Formulate country-specific policy objectives for the CDM in forestry;
2. Identify project types, e.g. planted forests, agro-forest practices, urban tree
plantations that match these policy objectives and rank by priority;
3. For all contemplated project types, identify a range of feasible and optimal
parameter values for height and crown cover;
4. From local knowledge or geographic information systems, delineate project areas
that appear promising ex ante;
5. For these areas, collect relevant data, e.g. land-use patterns, characteristic height
and canopy cover values for classes of prior vegetative cover; ownership sizes,
sequestration potential, opportunity costs, product markets, social and
environmental conditions and local stakeholder attitudes;
6. Based on this assessment, identify prime areas for afforestation and reforestation
CDM projects and rank land cover classes to be converted by priority;
7. For these classes, identify parameter values that would preclude the existing
vegetative cover from being considered as “forest” ex ante;
5
In many cases, there were several differentiated definitions for specific forest types, e.g. dense forests, open
forest, forest-grasslands, or degraded forests. In these cases, thresholds were combined following personal
judgement.
6
8. Step 7 and 3 will indicate infeasible combinations of prior conditions and desirable
project types; general guidelines ( Appendix I) and methods for this crucial step
are described below;
9. Establish feasible prior conditions-project type combinations and chose optimal
parameter values for height and crown cover which maximize eligibility;
10. In a separate step, select minimal area of a “forest”, based on parcel size
distribution of ownerships
Currently, detailed information on the prior state of vegetative cover, such as canopy cover,
tree heights, forest fragmentation or parcel sizes is missing for large parts of the world
(Indonesia, 2002). Remote sensing with a resolution necessary for quantifying such measures
is economically infeasible (Dutschke 2002). Information on growth, yield, wood density,
biomass functions and sequestration potential is also lacking for many prospective project
types (Rakonczay 2002). Hence, expert judgment may have to substitute for firm data;
decisions will often have to be taken under uncertainty.
While mathematical decision models for an optimal choice of minimum parameter values in
the forest definition exist, e.g. linear or dynamic programming, Bayesian decision analysis or
simulation, data requirements for these models are unlikely to be met, and the effort of
establishing them might be disproportionate to prospective benefits for countries.
The following elaborates individual steps in the decision process described above, illustrates
some policy objectives, prior land cover classes, and common project types and simple
graphical techniques to arrive at an optimal solution.
Furthering national policy objectives with the CDM
International climate change agreements and the CDM will not be a panacea for funding
problems in land use and forestry, but they can provide new incentives to employ
afforestation and reforestation as a tool to further sustainable development. In this context,
countries may see a role for the CDM in achieving a wide range of policy goals. They might
seek to maximize eligible area for the CDM to expand the resource base for domestic forest
industries through plantations or rehabilitation, or for renewable wood energy. They might
perceive CDM forestry projects not only under the aspect of climate change mitigation, but
also as a means for adapting to climate change, enhancing development and food security for
smallholders or rural communities. Under the rules for the CDM, afforestation could well
serve the goal of increasing livelihoods and human well-being in urban centres or agricultural
landscapes. The protective function of new forests for biodiversity, soils and water might be
as important in some landscapes as the carbon they accumulate.
Accommodating prior site conditions
The Global Forest Resource Assessments 2000 (FAO, 2001) provides a distribution of prior
land cover classes for the pan-tropics and permits establishing actuarial transition probabilities
to plantations (Table 1).
7
Table 1: Transitions to plantations 1990 -2000 for the pan-tropics (FAO, 2001)
Transition to plantation from land
cover class
Million ha in the
year 2000
%
1.9
37
0
0
Fragmented forest
0.2
4
Long fallow
0.2
3
Short fallow
0.4
7
Shrubs
0.2
4
Other land cover classes
2.3
45
Closed forest
Open forest
In the past, plantations have replaced mostly closed forests or non- forestry land uses. The
stringent restrictions imposed by CDM rules and modalities prevent closed forests, and most
open forests and long fallows from being converted to plantations. Only the following land
cover classes could conceivably furnish sites for afforestation and reforestation under CDM:
Fragmented forests: In 2000, FAO identified a total of 223 Million ha of such forests in the
pan-tropics. Individual fragments of the original forest are likely to exceed any threshold for
height. Forest fragments in human-dominated landscapes, tend to be below one ha in size
(Laurance, 2005), Given a goal to render such lands eligible for CDM projects, e.g. to
reconnect scattered parcels of remaining forests, a low minimum area could render interstices
between fragments eligible for projects and preserve existing fragments. Alternatively, a large
assessment unit combined with a high minimum crown cover might keep average canopy
cover over the entire project area below a high threshold value and render the entire area
eligible for A&R.
Open forests with crown cover below 30%: FAO defines open forests as having a canopy
cover of between 10% and 40% and inventoried 290 Million ha of them by the year 2000 in
the pan-tropics. Conceivably, some open forests, e.g. very severely degraded natural forests,
could be eligible for A&R projects, if countries use the upper limit of the available range to
define a forest and if these lands would not naturally revert to the forest stage.
Short fallows: In 2000, FAO surveyed 147 Million ha of such lands in the tropics. With the
short return time to agriculture, the fraction of parcels with woody vegetation is smaller than
in long fallow lands, and fields predominate. The system might conceivably not be considered
as forest or reverting to forest, if average height of regenerating tree species or average
canopy cover over an assessment unit are expected to remain below thresholds. In most cases,
however, individual parcels of these lands will have been forest after 1990 and will not be
eligible for the CDM.
Shrubs: In 2000, shrub lands made up 151 Million ha in the tropics alone and 1300 Million
ha globally. Such lands are dominated by woody perennial plants, generally more than 0.5 m
and less than 5 m high at maturity without a definite crown. Land may contain scattered and
clustered trees. To exclude the risk that such lands could be forests under the Protocol,
thresholds for crown cover of trees and minimum height should be at the upper boundary.
Clearly, shrub lands on better sites could present a large potential source of lands for the
CDM.
8
Other land cover with less than 10% woody vegetation: Irrespective of the minimum
parameter values chosen for area, crown cover, and height, there is no risk that land under this
category might be forests ex ante. Parameter value choice can focus on prerequisites for
desirable project types instead. In 2000, such lands amounted to 1017 Million ha in the pantropics.
Facilitating project types
Major potential project types include those listed below (see Figure 1).
Agroforestry
Agroforestry practices are diverse in their woody and herbal components, management and
biophysical settings (Montagnini and Nair, 2004; Appendix II), but total system productivity
will depend strongly on crown-cover (Kant 2005).
Introducing a small tree component into denuded, purely agricultural landscapes to establish
some agroforestry practices (Pancel, 1993), such as protein banks for animal fodder,
multipurpose trees on croplands, trees on rangelands, estate crop trees with pasture, or trees
for soil conservation and reclamation would necessitate a relatively low canopy cover
threshold in order to avoid productivity losses for the agricultural components of the systems.
Other agroforestry practices with a more substantial woody component, such as Taungya
systems, tree gardens, or some estate crop combinations operate optimally with higher tree
stocking and would be feasible as afforestation projects even under prior conditions that
include scattered trees with a crown cover at or above 10%.
Agroforestry practices that produce fodder through browsing, fuel wood through coppicing, or
hedgerows and alley cropping practices employing smaller trees all require a low threshold
for height to qualify ex post as forests. Choosing a small minimum forest area would facilitate
smallholder participation.
For living fences, shelterbelts, windbreaks, woody hedgerows selecting a low strip width
would maximize eligibility, unless minimum crown cover requirements are met at the
assessment unit level.
Urban forestry
Urban forestry can take very diverse forms and shapes; possible project types for the CDM
could range from afforestation on relatively large surfaces, where the choice of the forest
parameters is not critical, to small scale plot, strip- and alley planting. To reach “forest” status
for the latter type of urban greening, thresholds for crown cover, minimum area and height
should be at the lower end of the ranges.
Productive plantations
In the most typical case, large productive plantations are established with fast-growing tree
species on shrub lands, short fallows, cropland or grazing land of good site quality.
Fragmented and severely degraded forests with low height and sparse crown cover which are
not expected to revert to high forests might also be converted. After conversion, the planted
forest will typically quickly attain 30% crown cover, and 5 m height. A high value for crown
cover (30%) and height should be chosen for the land to be eligible ex ante (Kant 2005).
9
Protective plantations
On the other hand, choosing a minimum crown cover and low heights at maturity might be
optimal for protective plantations close to the natural timberlines in mountains, under boreal
conditions (Köhl 2000), or in dry lands, where timber quality is not the concern. Such
protective plantations could be effective even if small; a low area threshold may therefore be
indicated.
Permanent tree crops and wood-energy plantations
A low value for minimum height may be necessary to ensure that bio-energy plantations or
agricultural systems involving crops from small tree are eligible. Oil-producing tree species
are often cut back severely in order to increase yields and to facilitate harvesting of the fruits.
Small-scale afforestation and reforestation
Projects under this special category, created by Parties specifically for low income individuals
and communities in support of rural livelihoods, could comprise a total area of from less than
100 ha up to 4000 ha and any project type described above. However, in order to facilitate
“bundling” of small-holder plots, minimum area needs to reflect prevailing parcel size
distributions. In many regions, small-holder farms could only participate if low area
thresholds are chosen.
Integrating prior site conditions and project types
Figure 2
Figure 3
45
10
40
8
30
height (m)
height (m)
35
25
20
15
6
height'
4
height''
10
2
5
0
0
0
10
20
30
40
50
60
% crown cover
70
80
90
100
0
10
20
30
40
50
60
70
80
% crown cover
prior stand and priority
posterior stand and priority
Figures 3 and 4: Comparing prior and posterior stand parameters; see text for explanation
Contrasting prior land cover characteristics with feasible parameter values for contemplated
project types helps to eliminate infeasible options and facilitates the choice of the lower
boundaries for crown cover and height, as illustrated in Figures 2 and 3.
Paired values of crown cover and height of typical existing stands for a hypothetical
landscape are plotted. Feasible values for height and crown cover of the contemplated
plantations are also plotted. Plot symbols express by their relative size the rank, value or
utility of prior and posterior stand types for policy goals. Vertical and horizontal dotted lines
10
represent minimum values. Numbering the quadrants formed by these intersecting lines
clockwise, lines should be chosen as to maximize the number (or area / utility) of existing
stand types in quadrants one, three, and four, while simultaneously maximizing the number
(area, value, utility) of contemplated stand types in quadrant two.
Figure 2 depicts prior stand characteristics of hypothetical heavily fragmented, degraded
remnants of a forest, and stand characteristics for productive plantations. A crown cover of
30% and a minimum height of 5 m would be the optimal choice maximizing available area
All plantations considered are feasible as forests.
Figure 3 represents a hypothetical intensively managed very short fallow landscape to be
utilized for sustainable agroforestry practices with a low component of trees. A minimal
crown cover of 30% and the largest possible value for height of 5 meter (height’) maximizes
eligibility of existing lands for the CDM. However, with those parameter values, only a single
agro forest practice, which is also not the most desirable (Quadrant 2), could be established on
eligible land. Selecting a lower thresholds for height of 2m (height’’) would not affect land
availability, but would increase availability of agroforestry options, include project types of
greatest utility and, therefore, represent a better choice.
Integrating parcel size distribution
Figure 4: Establishing parcel size distribution by the scarce data method
Cumulative frequency distribution of parcel size
ordered
percentile
parcel rank
of the
size
distribution
1
2
3
4
5
6
7
8
9
10
9.09
18.18
27.27
36.36
45.45
54.54
63.63
72.72
81.81
90.90
cumulative frequency (%)
0.01
0.03
0.04
0.50
0.60
0.60
0.80
1.20
2.00
4.50
100
80
60
40
20
0
0
1
2
3
4
5
parcel size (ha)
parcel size
cumulative frequency
minimum area 1 ha
The parameter value for the definition of minimum area may strongly influence participation
by rural people in areas where small-holders prevail. However, in many developing countries
records to establish a distribution of the size of individual holdings may not be readily
available. Expert estimates may substitute for objective data. An easily applied and quite
reliable sampling method is the “scarce data method” (Snedecor and Cochran, 1978; Schoene,
1983).
According to this method, the “k” th observation in a random sample of “n” values which
have been ordered by size is an estimator of the “k/n+1” th percentage of the underlying
cumulative frequency distribution. Even relatively few observations suffice to estimate the
11
underlying population distribution. Figure 4 provides an example using the distribution of
parcel sizes to decide on minimum forest area. For a minimum area of 1 ha, 62% of smallholders could not participate in the CDM; obviously a lower threshold should be chosen if
smallholders are to benefit from the CDM.
What parameter values have NAI countries selected up to the present?
Figure 5 provides the distributions of values reported for these parameters to the UNFCCC
(UNFCCC, 2006), with FAO values given as a reference.
12
16
12
FAO
10
8
6
4
FAO
10
8
6
4
0
10
20
30
minimum crown cover (%)
0
8
6
4
2
2
2
FAO
10
Number of countries
12
Number of countries
Number of countries
14
2.5
3.0
3.5
4.0
4.5
minimum height (m)
5.0
0
0.0
0.1 0.2
0.3
0.4
0.5
0.6
0.7
minimum area
Figure 5: Distributions of minimum parameter values chosen by NAI countries by October
2006, compared to the FAO definition as a reference.
By choosing the minimum crown cover at the upper end of the range, reporting countries have
tried to eliminate the risk of prior vegetation being considered as forests and ineligible for the
CDM. China, the Dominican Republic and Nicaragua are exceptions.
A minimum height of below 5 m indicates that other countries, Albania, China, Vietnam
Yemen, Niger, Nicaragua, and Mali, may consider also project types other than highly
productive plantations to fulfil their policy goals for the CDM.
Less than half of all reporting countries, among them India or China, chose a small minimum
area, apparently addressing prevailing ownership patterns and policy goals.
12
1.0
Conclusions
Non-Annex I countries interested in hosting forestry CDM projects must choose parameter
values for the definition of forest from ranges given under the rules and modalities of the
CDM. Minimum crown cover, minimum height, and minimum area are required. Good
practice calls for also selecting a minimum strip width. Some terms, such as “tree”, “height”,
“temporarily unstocked state”, “expected return to forest”, and “promotion of natural seed
sources” remain ambiguous. Together with the significant role of spatial assessment units
these issues might be clarified by the Executive Board of the CDM.
Neither transposing existing national definitions of forest, nor resorting to the FAO definition
will alleviate the need for a rational national choice. The selection will affect eligibility,
extent, prospects, and impacts of forestry CDM projects. Parameters should match prior site
conditions, allow for desirable project types, and facilitate national policy goals. Optimal
values will usually vary by country.
Available data on existing country definitions, historic transitions to planted forests, and
details of possible project types provide a decisions background. A stepwise decision
algorithm and relatively simple graphical techniques may facilitate the urgent choice of
optimal parameter values for the forest definition by NAI countries.
13
References
Bellefontaine, R.2005.Régéneration naturelle à faible coût dans le cadre de l’aménagement
forestier en zones tropicales sèches en Afrique. Institute des Sciences de
l’Environnement. Université de Quebec, Montréal.
Dutschke, M. 2002. Sustainable forestry investment under the Clean Development
Mechanism: the Malaysian case. HWWA Discussion Paper 198.
FAO.2000. FRA 2000: On definitions of forest and forest change. Forest Resource
Assessment. Working Paper 33. Rome.14 p.
FRA 2000a.Comparison of forest area and forest area change estimates derived from FRA
1990 and FRA 2000. Forest Resources Assessment Working Paper 59
FAO. 2001. Global Forest Resource Assessment 2000. FAO Forestry Paper 140, FAO, Rome.
FAO. 2005. Global forest resource assessment. http://www.fao.org/forestry/site/24690/en
FAO, Rome.
FAO. 2006. Definitions related to reducing emissions from deforestation in developing
countries. FAO Forestry Paper. In print.
Indonesia. 2002. Views from Parties on issues related to modalities for the inclusion of
afforestation and reforestation project activities under the Clean Development
Mechanism in the first commitment period. FCCC/SBSTA/2002/MISC.22/Add.3
Submission to the UNFCCC Secretariat.
IPCC. 2003.Good practice guidance for land use, land-use change and forestry. IGES.
Hayama, Japan.
Helms, J.A. The dictionary of forestry. Society of American Foresters. Bethesda. Md. USA.
210p.
Kant, P. 2005. Definition of forests under the Kyoto Protocol: choosing appropriate values
for crown cover, area and tree height for India. Indian Forester.
Köhl, M. 2000. Reliability and comparability of TBFRA-2000 results. Pages 27-41 in UNECE/FAO Forest Resource Assessment, New York.
Montagnini, F., and P. K. R. Nair. 2004. Carbon sequestration: an underexploited
environmental benefit of agroforestry systems. Agroforestry Systems 61:281-295.
Laurance, W. F. 2005.Forest-climate interactions in fragmented tropical landscapes. In:
Tropical forests and global atmospheric change. Y.Malhi and O.Phillips, eds.p.31-38
Nair, P. K. R., editor. 1989. Agroforestry systems in the tropics. Kluwer Academic
Publishers.
15
Neeff, T. 2005. Ainda há incentivos para o desflorestamento no Mecanismo de
Desenvolvimento Limpo? Boletim Brasileiro de Meteorologia.
Pancel, L. 1993. Tropical Forestry Handbook. Vol. I and II. Springer. Berlin
Prodan, M. 1965. Holzmesslehre. Sauerlaender. Franfurt.644 p.
Rakonczay, Z. 2002. Biome-specific forest definitions. FCCC/TP/2002/1 Report to the
UNFCCC Secretariat.
Schoene, D.1983. Eine Alternative zur Vollkluppung. Allgemeine Forstzeitschrift 22: 571572
Snedecor, G.W. and W.G.Cochran. 1978. Statistical methods. Iowa State University Press,
Ames, Iowa.
UNFCCC. 2005. Website of the UNFCCC. http://unfccc.int August 2005.
UNFCCC.2006. Website of the UNFCCC. http://cdm.unfccc.int/DNA
Verchot, L. V., R. Zomer, O. van Straaten, and B. Muys. 2005. Implications of countrylevel decisions on the specification of crown cover in the definition of forests for land
area eligible for afforestation and reforestation activities in the CDM. to be
determined.
16
Appendix I
Criteria for the choice of high and low parameter values for the CDM forest definition.
Criteria for choosing low values
Criteria for choosing high values
Agroforestry with a
low tree component
< 30%
Main project type
productive
plantations
Plantation in dry
forests with crown
cover < 30%
Plantation in
mountainous, boreal
or dry forests with
height < 5 m
Permanent crops or
bio-energy
plantations at height
<5m
Large areas with
croplands/pasture
lands with crown
cover > 10 %?
Crown cover
Restoration of
degraded forests at
crown cover > 10%?
Height at maturity
Area
Restoration of
degraded shrubland
at height < 5 m?
Urban forestry an
option?
Highly fragmented
landscapes with
patches < 1 ha?
Activities on areas
< 1 ha
17
Appendix II
Agroforestry practices6 in small-scale A&R7
System
Practice
Combination
Agrisilvicultural
systems
1. Improved fallow
trees planted during
non-forest phase, if
land not expected to
revert to forest
2. Taungya
crops during tree
seedling stage
3. alley cropping
trees in hedges, crops in
alleys
4. tree gardens
multispecies, dense,
mixed
5. Multipurpose trees on
cropland
trees scattered,
boundaries
6. estate crop
combinations
7. Homegardens
multi-storey mixes
around homes
Agrosilvopastoral;
systems
9. shelterbelts,
windbreaks, live hedges
around farmland plots
10. Fuel wood production
firewood species
around cropland plots
11. Trees on rangelands
scattered trees
12. Protein banks
trees for protein-rich
cut fodder
13. Estate crops with
pasture
Example cattle under
coconut palms
14. Homegardens with
animals
around homes
15. Multipurpose woody
hedgerows
trees for browsing,
mulch, soil protection
16. Aqua forestry
trees lining ponds
6
Pancel, 1993
7
W: woody species; a: animals; h: herbaceous(crop) species
18
w: fast growing
h: agricultural crop
w: plantation species
h: agricultural crops
w: coppice trees
h: crops
w: vertical structure, fruit
trees
h: shade tolerant
w: multipurpose trees
h: crops
w: coffee, coconut, fruit
trees
h: shade tolerant
w: fruit trees
h: crops
w: multipurpose fruit trees
h: crops
8. trees in soil
conservation, reclamation
Silvopastoral
systems
Components
w: trees
h: crops
w: firewood species
h: crops
w: multipurpose, fodder
f: present
a: present
w: leguminous trees
h: present
a: present
w: estate crops
F: present
a: present
w: fruit trees
a : present
w: coppicing fodder trees
a, h: present
w: leaves forage for fish